Minimax TD-Learning with Neural Nets in a Markov Game.
Fredrik A. DahlOle Martin HalckPublished in: ECML (2000)
Keyphrases
- neural nets
- td learning
- evaluation function
- minimax search
- game tree
- temporal difference
- back propagation
- feed forward
- neural network
- monte carlo
- markov chain
- artificial neural networks
- function approximation
- game playing
- reinforcement learning
- learning tasks
- reinforcement learning algorithms
- active learning
- average reward
- policy evaluation
- step size
- model free
- multi step
- data mining